D-mixing and indirect CP violation measurements at LHCb
CCKM2014October 3, 2018
D-mixing and indirect CP violation measurements at LHCb Silvia Borghi School of Physics and Astronomy,The University of Manchester, Manchester, UK
The LHCb experiment collected during run I the world’s largest sampleof charmed hadrons. This sample is used to search for CP violation in charmand for the measurements of D mixing parameters. The measurement ofthe D − D mixing parameters and the search for indirect CP -violationin two-body charm decays at LHCb experiment are presented.PRESENTED AT Presented at the 8th International Workshop on the CKMUnitarity Triangle (CKM 2014)Vienna, Austria, September 8-12, 2014 On Behalf of the LHCb Collaboration a r X i v : . [ h e p - e x ] D ec Introduction
The charm sector is a promising field to probe for the effects of physics beyond theStandard Model (SM). Flavour mixing in the charm sector is now well established [1].In the SM [2, 3], the CP violation in charm transitions is expected to be small, withasymmetries up to few O (10 − ), while it can be enhanced by contribution from NewPhysics [4].The LHCb experiment is dedicated to the study of b and c flavour physics. Theabundance of charm particles produced in LHC offers an unprecedented opportunityfor high precision measurements in the charm sector, including measurements of CP violation and D − D mixing.The results of search for indirect CP violation and the measurements of the mixingparameters in two body hadronic D charm decays are presented here. CP violation with D → K + π − decays The flavour mixing occurs because the mass eigenstates ( | D , (cid:105) ) are linear combinationsof the flavour eigenstates and they can be written as linear combinations of theflavour eigenstates | D , (cid:105) = p | D (cid:105) ± q | D (cid:105) , with complex coefficients p and q whichsatisfy | p | + | q | = 1. The mixing parameters are defined as x ≡ ( m − m ) / Γ and y ≡ (Γ − Γ ) / (2Γ), where m , m , Γ and Γ are the masses and the decay widths forD and D , respectively, and Γ = (Γ + Γ ) /
2. The phase convention is chosen suchthat CP | D (cid:105) = −| D (cid:105) .The first evidence of D − D oscillation was reported in 2007 by BaBar [5] and Belle[6]. Now, the mixing in the charm sector is well established with the first observationby a single measurement with greater than 5 standard deviation significance at theLHCb experiment [7], confirmed by CDF [8] and Belle [9].At the LHCb experiment, the charm mixing parameters are determined by thedecay-time-dependent ratio of D → K + π − (called wrong sign, WS) to D → K − π + (called right sign, RS) decay rates. The RS decay rate is dominated by Cabibbofavoured (CF) amplitude. The WS rate arises from the interfering amplitudes ofthe doubly Cabibbo-suppressed decay (DCS) and the CF decay following D − D oscillation. Assuming no CP violation and small mixing parameters ( | x | and | y | (cid:28) R ( t ) ≈ R D + (cid:112) R D y (cid:48) tτ + x (cid:48) + y (cid:18) tτ (cid:19) where x (cid:48) = x cos δ + y sin δ , y (cid:48) = y cos δ − x sin δ , R D is the ratio of suppressed-to-favoured decay rates, δ is the strong phase difference between the DCS decays andthe CF decays (cid:0) A ( D → K + π − ) / A ( D → K − π + ) = −√ R D e − iδ (cid:1) .1 ] - [ y ' LHCb
CPV allowed (a) D D ] -3 [10 x' -0.1 0 0.1 0.2 No direct CPV (b) D D -0.1 0 0.1 0.2 No CPV (c)
Figure 1: Two-dimensional confidence regions in the ( x (cid:48) , y (cid:48) ) plane obtained a) withoutany restriction on CP violation, b) assuming no direct CP violation and c) assuming CP conservation. The solid (dashed) curves in a) and b) indicate the contours of themixing parameters associated with D ( D ) decays. The solid, dashed and dottedcurves in c) indicate the contours of CP -averaged mixing parameters at 68 . . .
7% C.L.. The best-fit value is shown with a point [10].One can search for CP violation in the charm sector by comparing the time-dependent ratios evaluated separately for the two flavours ( D and D ). A differencein the R D parameter between D and D would be a sign of direct CP violation.While a difference in x (cid:48) and y (cid:48) parameters would imply an indirect CP violationcontribution ( | q/p | (cid:54) = 1 or φ = arg ( q/p ) (cid:54) = 0). The data are fit considering threescenarios: 1) assuming CP conservation 2) allowing only indirect CP contribution; 2)allowing both direct and indirect CP contributions.The full data sample with a total integrated luminosity of 3 fb − is used to performthese measurements. The analysis is performed on D ∗± → D π ± decays to allow thedetermination of the flavour of the neutral D meson at production.In the scenario 1) the mixing parameters are measured to be R D = (3 . ± . ± . · − , y (cid:48) = (4 . ± . ± . · − and x (cid:48) = (5 . ± . ± . · − wherethe first uncertainty is statistical and the second systematic. In the scenario 2) and3), the magnitude of | q/p | is constructed. The magnitude of q/p is determined to be0 . < | q/p | < .
24 and 0 . < | q/p | < .
52 at 68 .
3% and 95 .
5% C.L., respectively [10].The results of the mixing parameters measurements and of the confidence regions areshown in Fig. 1 for the three scenarios. They are compatible with CP conservation andprovide the most stringent bounds on the parameter | q/p | from a single experiment.2 Indirect CP violation in 2-body D decays to CP eingenstates A measurement of the indirect CP violation in D mixing can be performed in the studyof two-body hadronic charm decays to CP eigenstates ( D → K + K − or D → π + π − ).It can be evaluated by the asymmetry in the effective lifetimes ( τ ) of flavour-taggeddecays and it can be expressed by the following equation with the assumption ofnegligible direct CP -violation contribution: A Γ = τ ( D → f ) − τ ( D → f ) τ ( D → f ) + τ ( D → f ) ≈ A m y cos φ − x sin φ where A m is defined by | q/p | ± ≈ ± A m . A measurement of A Γ differing significantlyfrom zero is a manifestation of indirect CP violation as it requires a non-zero valuefor A m or φ .The measurement of A Γ at LHCb is performed using 1 fb − of data from a sampleof pp collisions at a centre-of-mass energy of 7 TeV collected in 2011. The flavour atproduction is again determined using neutral D mesons from D ∗± → D π ± decays.The main selection is applied at the trigger level on the momentum, PID and impactparameter (IP) of the D daughters. The trigger event selection causes a bias of theproper-time distribution. Thus, an acceptance correction is needed for the evaluationof the effective lifetime. The acceptance is determined using a data driven method,the so-called swimming algorithm [11, 12].The fit to determine the effective lifetime is performed independently for eachflavour tag and each decay mode. The signal yield are extracted from simultaneousunbinned likelihood fits of the D invariant mass and of the difference between D ∗ and D masses, ∆m, to distinguish the different background contributions. Charm mesonsproduced in b -hadron decays, secondary charm, have larger impact parameter withrespect to the primary vertex than the prompt candidates as a secondary D does notusually point back to the primary vertex. This additional background is subtracted inin a simultaneous fit of the proper-time and ln ( χ IP ) distributions. The χ IP is definedas the difference in χ of a given primary interaction vertex reconstructed with andwithout the considered particle. Fig. 2 shows an example of the proper-time projectionfor D → K + K − and D → π + π − decays for one data set.The analysis method is validated on a control sample of CF D → K − π + decays,where the lifetime asymmetry is determined to be consistent with zero in accordancewith the expectation. The resulting values of A Γ for the two final states [13] are: A KK Γ = ( − . ± . ± . · − A ππ Γ = (0 . ± . ± . · − n t r i e s / ( . p s ) DataFitSignalSecondary s p Prompt rnd. s p Sec. rnd. p + p - K fi D + p - K + K fi +s DComb. bkg
LHCb t [ps] P u ll -505 E n t r i e s / ( . p s ) DataFitSignalSecondary s p Prompt rnd. s p Sec. rnd. Comb. bkg
LHCb t [ps] P u ll -505 t [ps]1 2 3 ) ) / n ( D D n ( ) ) / n ( D D n ( Figure 2: Lifetime fit projection of D → K + K − decays (on the top left) and of D → π + π − decays (on top right) and corresponding pull plot, for one data set. Theratio of D to D data and fit model for decays to KK (on the left bottom) and π + π − (on the right bottom) for all data, respectively [13].where the first uncertainty is statistical and the second systematic. Both resultsare consistent with zero, showing no evidence for indirect CP violation. They areconsistent with and more precise than previous determinations from other experiments[1]. Amongst the several sources of systematics considered, the main ones are due tothe decay-time acceptance correction and due to the background description. The mixing in charm sector is now well established, while searches for indirect CP violation yield results consistent with CP conservation. Further measurements areongoing at LHCb using the data set collected during Run I and others will followwith the upcoming Run 2, allowing to explore the charm sector with unprecedentedprecisions. Later, the LHCb Upgrade is expected to collect 50 fb − of integratedluminosity, allowing to reach precisions down to 0 . · − for A Γ and O (10 − , − )for ( x (cid:48) , y (cid:48) ). 4 eferences , 501 (1989).[3] M. Bobrowski, A. Lenz, J. Riedl and J. Rohrwild, JHEP
009 (2010).[4] Y. Grossman, A. L. Kagan and Y. Nir, Phys. Rev. D , 036008 (2007).[5] BaBar Collaboration, B. Aubert et al., Phys. Rev. D112